JP2008083816A - Travelling control system for automobile and vehicle controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To highly accurately recognize an on-road object in an early stage for vehicle control. <P>SOLUTION: This vehicle controller 100 is provided with: a first road environment information acquisition means for acquiring the information of a road environment by recognizing the peripheral on-road object of a vehicle 10; a second road environment information acquisition means for acquiring the information on the road environment by recognizing an on-road object which is different from the on-road object recognized by the first road environment information acquisition means; and a road environment recognizing means for recognizing road environment ahead of its own vehicle according to the first road environment information acquired by the first road environment information acquisition means and the second road environment information acquired by the second road environment information acquisition means. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an automobile travel control system and a vehicle control device.

  A technique for determining whether a toll gate exists in front of a vehicle is known.

  For example, the technique disclosed in Patent Document 1 calculates a vehicle position, obtains map data around the vehicle position from a map database, and compares the vehicle position with the map data to advance the vehicle in the traveling direction. Determine whether a toll booth exists.

In addition, the technique disclosed in Patent Document 2 uses a radar to scan a toll gate, and the number of reflection points detected is different from the number of reflection points detected when another front object is scanned. A method for determining a toll gate by using the fact that it becomes very large is described.
JP 2001-134794 A JP 2004-69328 A

  Features are locations on the road that require vehicle control, such as tollgates, rampways, interchanges, parking areas, curves, railroad crossings, and intersections. It is required to recognize these features with high accuracy at an early stage. Thereby, it is possible to determine whether there is a tollgate or the like and to control the speed of the vehicle or to notify the occupant.

  For example, the automatic toll collection system (ETC: Electronic Toll Collection System) has become widespread, so the general toll booth and ETC can be identified to change the speed when passing through the toll booth, and the occupant is informed of the vehicle's driving status It is necessary to do. For this reason, it is effective to recognize the type of toll booth, which is a feature, with high accuracy at an early stage.

  However, the techniques described in the above-mentioned patent documents do not sufficiently meet such demands.

  First, the technique disclosed in Patent Document 1 has insufficient accuracy. When a new toll booth is constructed, the toll booth needs to be reflected in the map database. However, frequent updating of the map database results in a significant cost increase, including on-site surveys. An old map database that has not been updated cannot fully recognize features.

  On the other hand, the technique disclosed in Patent Document 2 is delayed in recognition. Recognition by a single radar has physical limitations in terms of irradiation distance and scanning performance. For example, when speed control is performed, it is necessary to mount a radar capable of irradiating a long range in order to start deceleration much before a toll gate. In addition, it is necessary to greatly improve the scanning range in order to accurately control the speed when passing through the toll gate. If such equipment is installed, a significant increase in cost will occur. Therefore, in the current recognition of a narrow range, recognition is not performed until the feature is approached.

  Accordingly, the main object of the present invention is to recognize a feature with high accuracy at an early stage for vehicle control.

  In order to solve the above-mentioned problem, the present invention recognizes the first road environment information acquisition means for recognizing features around the own vehicle and acquiring road environment information, and the first road environment information acquisition means. Second road environment information acquisition means for recognizing a feature different from the feature and acquiring road environment information; first road environment information acquired by the first road environment information acquisition means; and the second Road environment recognition means for recognizing the road environment ahead of the host vehicle in accordance with the second road environment information acquired by the road environment information acquisition means. Other means will be described later.

  According to the present invention, before the second road environment information acquisition unit recognizes the feature itself, the first road environment information acquisition unit recognizes a different feature (for example, a road sign to the feature). Therefore, the feature can be recognized with high accuracy at an early stage. Therefore, speed control without a sense of incongruity to the driver and notification to the passenger are realized.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a state transition diagram showing feature recognition. This state transition diagram shows four states.

  The state of 101a during GPS (Global Positioning System) recognition is a state in which the distance between the vehicle position acquired by GPS and the position of the target feature is calculated. When this distance is equal to or smaller than a predetermined value, it is assumed that the feature is approaching, and the state transits to the state 101b during recognition of the road sign to the feature.

  The state of the road sign recognition 101b to the feature is a state of recognizing the road sign to the feature from the image data taken by the camera (imaging device) mounted on the vehicle. The road sign to the feature is a sign that visually informs the driver that the feature is present at the destination of the road. For example, it is described as “There is a △ junction at 2 km ahead”. As a result, the vehicle control device (details will be described later) finds that a feature called △ Δ junction exists 2 km ahead. When the calculated distance to the feature is equal to or less than a predetermined value, it is assumed that the feature is approaching, and the state transitions to the state of feature recognition 101c.

  The state of feature recognition 101c is a state of recognizing a feature from image data taken by a camera mounted on the vehicle. When the calculated distance to the feature is equal to or less than a predetermined value, it is assumed that the feature has been reached, and the state transitions to the state 101d during vehicle control.

  The state 101d during vehicle control is a state in which vehicle control corresponding to the feature is being performed. For example, when the feature is a junction, vehicle control is performed such that the junction is turned to the left. When the vehicle control is finished, the vehicle moves to the next feature, so that the state again transits to the state of GPS recognition 101a.

  As described above, the state of the road sign recognition 101b to the feature is set between the state of GPS recognition 101a and the state of feature recognition 101c. In the state of the road sign recognition 101b to the feature, the vehicle position may be corrected to the road sign position. Thereby, the precision of a vehicle position improves.

  Further, in the state of 101b during the recognition of the road sign to the feature, the vehicle control that is a preparation for the vehicle control performed in the state of the vehicle control 101d may be performed. As a result, smooth vehicle control can be performed with a margin as compared with a method in which vehicle control is suddenly performed after arrival at a feature.

  FIG. 2 is an explanatory diagram showing a photographed image of a front camera attached to the front of the vehicle. Hereinafter, a toll gate will be described as an example of the feature. The toll gate recognition result is used for vehicle speed control and passenger notification.

  FIG. 2A is an explanatory diagram showing an example of a road sign to the feature recognized during 101b (see FIG. 1) during recognition of the road sign to the feature. A guardrail 354 is installed on the left side of a toll road 351 such as an expressway. The road sign 352 indicates a distance (for example, 1 km) to the toll gate. A road sign 353 indicates the location of the ETC gate. The road sign 352 and the road sign 353 are recognized as road signs to a toll gate, for example, by detecting their characters, colors and shapes.

  In addition, the peripheral image of the road sign 353 is stored in advance, and the road sign 353 is recognized by pattern matching (recognizing the size of the sign) between the stored image and the image acquired in real time from the front camera. Also good. Furthermore, it is also possible to obtain information on the toll gate in front of the vehicle (the number of lanes and the position of the ETC gate) according to the number of lanes indicated on the road sign 353.

  FIG. 2B is an explanatory diagram illustrating an example of a feature recognized during the feature recognition 101c (see FIG. 1). A toll gate 602 is installed on a toll road 601 such as an expressway. As a method of detecting the toll gate 602, characters, colors, and shapes of the guide plate (general) 611, the guide plate (ETC) 621, and the guide plate (ETC / general) 625 may be recognized. As another detection method of the toll gate 602, the color and shape of the road marking (for general use, white line) 612 and the road marking (for ETC, white mark on blue) 622 may be recognized.

  In addition, the color and shape of the general booth 613 and ETC circuit breaker 623 may be recognized, or the color and shape of the signals 614 and 624 installed in the toll gate 602 may be recognized up to the toll gate 602. The distance may be calculated. In addition, the tollgate's surrounding images are stored in advance, and the pattern matching between the stored images and the images acquired in real time (recognizes the size of the information boards, road markings, booths and breakers, or gate shapes) May recognize the toll gate, or may recognize the road signs 612 and 622 using the rear camera.

  FIG. 3 is a block diagram showing an automobile system. This automobile system recognizes the road environment ahead of the vehicle, and controls the vehicle and notifies the occupant according to the recognized road environment.

  The vehicle 10 is equipped with an engine 5 and a transmission 6, and power generated by the engine 5 is input to the power split mechanism 7 via the transmission 6. The power input to the power split mechanism 7 is distributed to the propeller shafts 8 and 9 and transmitted to the tires 21 and 22. The torque of the engine 5 is adjusted (controlled) in accordance with a control signal from an engine ECU (Electric Control Unit) (not shown).

  Brakes 11 and 12 controlled by actuators such as boosters are attached to the tires 21 and 22, respectively, and a frictional force is generated on the tires 21 and 22 by a driver's brake operation or a control signal of a brake ECU (not shown). 10 braking force is adjusted (controlled).

  In addition, the vehicle 10 includes an antenna that receives image information (mainly latitude and longitude) using a front camera 1 and a rear camera 2 that detect image conditions by acquiring image information, and a GPS satellite 4. A GPS receiver 3 that obtains the latitude and longitude of the current position is mounted, and a vehicle control device 100 that communicates with these sensors (camera, antenna), engine ECU, brake ECU, and the like via a LAN (Local Area Network) is mounted.

  Next, the configuration of the vehicle control device 100 will be described.

  The vehicle control apparatus 100 includes a distance predicting unit 101, a road sign recognizing unit 102 (a sign recognizing unit), a distance estimating unit 103 (a sign recognizing unit), a toll gate recognizing unit 104 (a feature recognizing unit), and a distance correcting unit 105 (a ground recognizing unit). (Object recognition means). Each of these means is realized by a CPU (Central Processing Unit) of a computer (not shown) of the vehicle control device 100 reading and executing a program in a memory at a predetermined cycle. In the present embodiment, the “first road environment information acquisition unit” corresponds to the road sign recognition unit 102, the “second road environment information acquisition unit” corresponds to the toll gate recognition unit 104, and the “road environment information acquisition unit”. The “recognizing means” corresponds to the distance estimating means 103 and the distance correcting means 105.

  The distance predicting means 101 predicts the distance from the current vehicle position to the toll gate according to the GPS signal (information such as the latitude, longitude, altitude of the vehicle) from the GPS receiver 3 and the map database, and outputs the predicted distance. .

  The road sign recognition unit 102 recognizes and outputs a road sign to the toll gate according to the predicted distance output by the distance prediction unit 101.

  The distance estimation means 103 estimates the distance from the current vehicle position to the toll gate according to the predicted distance output by the distance prediction means 101 and the road sign information output by the road sign recognition means 102, Output the estimated distance.

  The toll gate recognition unit 104 recognizes the toll gate according to the estimated distance output by the distance estimation unit 103 and outputs toll gate information.

  The distance correction unit 105 corrects the distance from the current vehicle position to the toll gate according to the estimated distance output by the distance estimation unit 103 and the toll gate information output by the toll gate recognition unit 104.

  The road sign recognition unit 102 and the toll gate recognition unit 104 preferably use at least one camera (front camera 1 and rear camera 2). For example, two image processing logics (for example, a road sign recognition unit 102 and a toll gate recognition unit 104) are mounted on one image processing processor using an image captured by one front camera mounted in front of the vehicle. As a result, an increase in cost due to an increase in cameras can be reduced.

  FIG. 4 is a flowchart showing a state during GPS recognition.

  FIG. 4A is a schematic diagram showing the vehicle position on the map database. There are various features P1 to Pn-1 on the route from the departure point P0 to the destination Pn. The vehicle position Q is located between the departure place P0 and the feature P1.

  FIG. 4B is a flowchart showing the processing contents of the distance prediction unit 101.

  First, in S201, a GPS signal is read. Next, in S202, the map database is searched according to the GPS signal read in S201. In S203, the vehicle position Q on the map database is updated in accordance with the map database search result performed in S202.

  In S203, the vehicle position Q on the map database may be specified based on the GPS signal as described above, or the vehicle position on the map database may be specified by well-known autonomous navigation. . In the autonomous navigation, the distance from the starting point P0 to the current vehicle position Q is calculated using wheel speed, steering angle, acceleration sensor, and the like.

Next, in S204, a search process from the vehicle position Q to the nth feature (P1, P2,..., Pn) ahead is performed, and it is determined whether or not a toll gate exists in the searched feature. Do. If it is determined that a toll gate exists (S204, Yes), the distance from the vehicle position Q to the toll gate is predicted, and the result is updated to the predicted distance Lp (S205). Formula (1) shows a method for calculating the predicted distance Lp when a toll gate is present on the kth feature Pk. In Equation (1), the first term on the right side indicates the distance from the vehicle position Q to the feature P1, and the other terms indicate the distance between the features.
Lp = | Q−P1 | + | P1-P2 | + ... + | Pk−1−Pk | (where 1 ≦ k ≦ n) (1)

  On the other hand, if it is determined that there is no toll gate (S204, No), a predetermined value (for example, a maximum value with a margin of 5 km) is substituted for the predicted distance Lp (S206).

  As described above, the distance from the vehicle position Q to the toll gate P1 can be predicted by the distance prediction means 101 using the GPS signal and the map database.

  FIG. 5 is a flowchart showing a state during recognition of a road sign to a toll gate. The flowchart in FIG. 5A is executed by the road sign recognition unit 102.

  First, in S301, the predicted distance Lp output by the distance prediction unit 101 is read. Next, in S302, it is determined whether or not the predicted distance Lp is smaller than a predetermined value (for example, 1.2 km). If the predicted distance Lp is equal to or greater than the predetermined value (S302, No), it is determined that the vehicle is not approaching the toll gate, and the process ends. When it is determined that the predicted distance Lp is smaller than the predetermined value (S302, Yes), it is determined that the vehicle is approaching the toll gate and the process proceeds to S303.

  Next, in S303, the image of the front camera shown in FIG. 2A is read, and the process proceeds to S304. Next, in S304, road sign information is detected according to the image data read in S303, and the process ends.

  As described above, the road sign recognition means 102 can recognize the road sign and detect toll gate information. Further, by performing image processing only when the predicted distance Lp is within a predetermined range, the load on the image processing processor can be reduced.

  The flowchart in FIG. 5B is executed by the distance estimation unit 103.

  First, in S401, the predicted distance Lp output by the distance predicting unit 101 is read, and in S402, information on the road sign (sign information) to the toll gate output by the road sign recognizing unit 102 is read. Next, in S403, it is determined whether or not each road sign shown in FIG. 2A exists according to the road sign information read in S402.

  First, when there is no road sign (S403, No), the predicted distance Lp is substituted for the estimated distance Ls (S408), and the process is terminated. On the other hand, when the road signs 352 and 353 exist (S403, Yes), the estimated distance candidate value Lsb is updated (S404). In S404, the estimated distance candidate value Lsb may be calculated according to the distance to the toll gate shown in the road sign 352 (see FIG. 2A) detected as road sign information, or a predetermined database The estimated distance candidate value Lsb may be calculated from the position of each road sign (see FIG. 2A) stored in

Next, in S405, it is determined whether or not the estimated distance candidate value Lsb updated in S404 is smaller than the predicted distance Lp. If the determination is established (S405, Yes), the estimated distance Ls is updated (S406). In S406, a change amount limiting process as shown in Expression (2) is performed, and the estimated distance Ls gradually approaches the estimated distance candidate value Lsb. Here, Lsz is a value one cycle before the estimated distance Ls, and cDLs # is a variation limit value (constant).
Ls = Lsz-cDLs #
Limiter Ls ≧ Lsb (2)

If the determination is not established (S405, No), the estimated distance Ls is updated (S407). In S407, a change amount limiting process as shown in Expression (3) is performed, and the estimated distance Ls gradually approaches the estimated distance candidate value Lsb.
Ls = Lsz + cDLs #
Limiter Ls ≦ Lsb (3)

  As described above, by using the distance estimation means 103, the current vehicle position to the toll gate according to the predicted distance output by the distance prediction means 101 and the information of the road sign output by the road sign recognition means 102. The distance from the current vehicle position to the toll gate can be accurately estimated by using the road sign image information in addition to the map data.

  FIG. 6 is a flowchart showing a state during toll gate recognition. The flowchart in FIG. 6A is executed by the toll gate recognition means 104.

  First, in S501, the estimated distance Ls output by the distance estimating means 103 is read. Next, in S502, it is determined whether or not the estimated distance Ls is smaller than a predetermined value (for example, 100 m). If the estimated distance Lp is greater than or equal to a predetermined value (S502, No), it is determined that the toll gate cannot be recognized by the front camera 1, and the process is terminated. If it is determined that the estimated distance Ls is smaller than the predetermined value (S502, Yes), it is determined that the toll gate can be recognized by the front camera 1, and the process proceeds to S503.

  Next, in S503, the image of the front camera 1 shown in FIG. 2B is read, and the process proceeds to S504. In step S504, the toll gate information is detected according to the image data read in step S503, and the process ends.

  As described above, the toll gate recognition unit 104 can recognize the toll gate and detect toll gate information. Further, by performing image processing only when the estimated distance Ls is within a predetermined range, the load on the image processing processor can be reduced.

  The flowchart of FIG. 6B is executed by the distance correction unit 105.

  First, in S701, the estimated distance Ls output by the distance estimating means 103 is read, and in S702, the information on the toll gate output by the toll gate recognition means 104 is read. Next, in S703, it is determined whether there is a toll gate shown in FIG. 2B around the vehicle according to the toll gate information read in S702.

  If the toll booth does not exist (S703, No), the estimated distance Ls is substituted for the distance Lt (S708), and the process is terminated. If there is a toll gate (S703, Yes), the distance candidate value Ltb is updated (S704). In S704, the distance candidate value Ltb is calculated according to the information on the toll gate recognized by the toll gate recognition means 104. Alternatively, the distance candidate value Ltb may be calculated from the position of a toll gate (see FIG. 2B) stored in advance as a database.

Next, in S705, it is determined whether or not the distance candidate value Ltb updated in S704 is smaller than the distance Lt. If the determination is established (S705, Yes), the distance Lt update process (S706) is performed. In S706, a change amount limiting process as shown in Expression (4) is performed, and the distance Lt is gradually made asymptotic to the distance candidate value Ltb. Here, Ltz is a value one cycle before the distance Lt, and cDLt # is a variation limit value (constant).
Lt = Ltz−cDLt #
Limiter Lt ≧ Ltb (4)

If the determination is not established (S705, No), the distance Lt is updated (S707). In S707, a change amount limiting process as shown in Expression (5) is performed, and the distance Lt is gradually made asymptotic to the distance candidate value Ltb.
Lt = Ltz + cDLt #
Limiter Lt ≦ Ltb (5)

  As described above, by using the distance correction unit 105, the current vehicle position to the toll gate according to the estimated distance output by the distance estimation unit 103 and the toll gate information output by the toll gate recognition unit 104. The distance from the current vehicle position to the toll gate can be obtained with high accuracy by using the image information of the toll gate.

  FIG. 7 is a graph showing a case in which speed control is performed in front of the toll gate, and the passenger is notified when passing through the toll gate. A speed control unit and a notification unit (not shown) of the vehicle control apparatus 100 control a brake, a throttle, and the like, and execute the process shown in the graph of FIG.

  The horizontal axis of the graph indicates the vehicle position, and the vertical axis indicates the distance from the current vehicle position to the toll booth, the target speed TVSP, and the guidance flag fGUIDE. Here, the guidance flag fGUIDE is a flag for notifying the occupant of the current traveling state, and when fGUIDE is set, the occupant is notified using a voice guide or a warning. In FIG. 7, it is assumed that the exact location of the toll gate is point g.

  First, the case “1” (when only the predicted distance Lp is used) indicated by a broken line in the figure will be described. When the toll gate is detected by searching the map database at point a, the predicted distance Lp indicated by the broken line in the figure gradually decreases as the vehicle approaches the toll gate. Thereafter, when the predicted distance Lp becomes smaller than a predetermined threshold at point b, the target speed TVSP indicated by the solid line in the figure gradually decreases from the initial speed Vo in order to perform automatic deceleration before the toll gate.

  Here, when only the predicted distance Lp is used, an error occurs in the distance to the toll gate due to an error in the GPS signal or the map database, and the position of the toll gate is misrecognized as the point e. The target speed TVSP is set to converge to the final speed Vf at the point e, and the guidance flag fGUIDE is also set to be set at the point e. Therefore, the driver feels uncomfortable due to the end of the deceleration before the toll booth and the passenger discomfort due to the deviation of the voice guide and alarm timing.

  Next, the case “2” indicated by the dotted line in the figure (when the estimated distance Ls is used in addition to the predicted distance Lp) will be described. The control up to point b is the same as in case 1 described above, but when the predicted distance Lp becomes smaller than a predetermined value at point c2 in the figure, it is determined that there is a road sign to the toll gate, and the road sign recognition means 102 Image processing is started.

  Thereafter, when road sign information is detected at point c, the estimated distance Ls is calculated by the distance estimating means 103, and the position of the toll gate is corrected from point e to point f. With this correction, the target speed TVSP is also corrected to converge to the final speed Vf at the point f, and the guide flag fGUIDE is also corrected to be set at the point f. Since the distance to the toll booth calculated from the information on the road signs is less error than the distance to the toll booth calculated from the GPS signal and map data, these corrections improve the accuracy of speed control and notification timing to passengers Can be planned.

  Next, the case “3” indicated by the solid line in the figure (when the corrected distance Lt is used in addition to the predicted distance Lp and the estimated distance Ls) will be described.

  The control up to point c is the same as in the case “2” described above. However, when the estimated distance Ls becomes smaller than a predetermined value at point d2, it is determined that a toll gate is present and image processing by the toll gate recognition means 104 is performed. Be started. Thereafter, when the toll gate is recognized at the point d, the distance correcting unit 105 calculates the distance Lt, and the position of the toll gate is corrected from the f point to the g point.

  Along with this correction, the target speed TVSP is also corrected to converge to the final speed Vf at the g point, and the guide flag fGUIDE is also corrected to be set at the g point. The distance to the toll booth calculated from information such as the toll booth information board and road markings is less error than the distance to the toll booth calculated from the information on the road sign described above. The accuracy and the accuracy of the notification timing to the occupant are further improved.

  As described above, by determining the target value of speed control and the notification timing to the occupant according to the estimated distance Ls output by the distance estimating means 103 and the distance Lt output by the distance correcting means 105, the toll gate It is possible to reduce the driver's uncomfortable feeling due to the end of deceleration in front and the passenger's uncomfortable feeling due to the deviation of the voice guide and alarm timing.

  The speed control means may switch the target value of the speed control according to the type of the toll gate identified by the toll gate identification means for identifying the type of the toll gate from the information on the sign, the guide board, and the road marking. . For example, the driver's convenience can be improved by setting the final speed Vf to 0 km / h (stop) at a general toll booth and 20 km / h at an ETC to control the speed when passing through the toll booth.

  Further, in addition to speed control, lateral control can be performed according to the recognized road environment. For example, lane keeping control for maintaining the current travel lane may be performed by controlling the steering in accordance with road signs 612 and 622 (see FIG. 2B), and information on the recognized branch road In addition to route guidance by conventional navigation combined with map information, control for automatically guiding the route of the vehicle may be performed.

  The present invention described above can be widely modified without departing from the spirit thereof as follows.

  For example, the toll gate recognizing means 104 may identify the type of toll gate according to information transmitted from the ETC receiver (such as availability of ETC). For example, it is determined whether or not an ETC card is inserted by an ETC receiver. When an ETC card is inserted, guidance to the ETC lane is performed by voice guidance or the like, and the final speed Vf is set to 20 km / h. Set to.

  When the ETC card is not inserted, guidance to the general lane is performed by voice guidance or the like, and the final speed Vf is set to 0 km / h (stop). By using such a system, it becomes possible to prevent a vehicle without an ETC card from entering the ETC gate, thereby improving the convenience of the driver and eliminating traffic congestion near the toll gate.

  Various features can be used instead of the illustrated toll booth.

  An example of using a branch road as a feature will be described. For example, on branch roads such as rampways, interchanges and parking areas on highways, signs similar to the road signs shown in FIG. Is possible. At this time, the feature recognizing means recognizes a guide plate existing immediately before the branch road (for example, a green guide plate written as “P” in the parking area of the expressway).

  An example of utilizing a curve as a feature will be described. In a curve, it is possible to apply a sign recognition means by using a sign indicating a “forward sharp curve”. At this time, the feature recognition means recognizes the curve curvature and the relaxation curve before the curve from the white line.

  An example of utilizing a railroad crossing as a feature will be described. In a crossing, the sign recognition means of the present invention can be applied by using a sign indicating "front crossing". At this time, the feature recognition means recognizes a crossing breaker and a red flashing light.

  An example of using an intersection as a feature will be described. At the intersection, a guide board (a blue background guide board with directions and distances written in white letters) is installed on the road. As a means, the guide plate immediately before the intersection is recognized.

  FIG. 8 is an explanatory diagram illustrating a case where the label recognition is performed a plurality of times. First, when the sign recognition means recognizes the first sign “2 km ahead ○ Δ branch point”, vehicle control “change to left lane” is performed. Next, when the sign recognition means recognizes the second sign “500 m ahead ○ Δ branch point”, vehicle control “decelerate to 60 km / h” is performed. When the feature recognizing means recognizes the third sign “◯ △ branch point”, the vehicle control “proceeding in the left direction” is performed. As described above, the vehicle control is performed step by step corresponding to each of the plurality of signs, so that the burden on the vehicle occupant can be reduced and the ride comfort is improved.

It is a state transition diagram which shows the feature recognition regarding one Embodiment of this invention. It is explanatory drawing which shows the picked-up image of the camera mounted in the vehicle regarding one Embodiment of this invention. It is a block diagram which shows the motor vehicle system regarding one Embodiment of this invention. It is a flowchart which shows the state in GPS recognition regarding one Embodiment of this invention. It is a flowchart which shows the state during the road sign recognition to the feature regarding one Embodiment of this invention. It is a flowchart which shows the state in the feature recognition regarding one Embodiment of this invention. It is a graph which shows the case where speed control is performed before the toll gate regarding one Embodiment of this invention, and it alert | reports to a passenger | crew when passing a toll gate. It is explanatory drawing which shows the case where the label | marker recognition of one time regarding one Embodiment of this invention is performed.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Front camera 2 Rear camera 3 GPS receiver 4 GPS satellite 5 Engine 6 Transmission 7 Power split mechanism 8, 9 Propeller shaft 10 Vehicle 11, 12 Brake 21, 22 Tire 100 Vehicle control device 101 Distance prediction means 102 Sign recognition means 103 Distance Estimating means 104 Feature (tollgate) recognizing means 105 Distance correcting means

Claims (15)

First road environment information acquisition means for recognizing features around the vehicle and acquiring road environment information;
Second road environment information acquisition means for acquiring a road environment information by recognizing a feature different from the feature recognized by the first road environment information acquisition means;
A road that recognizes the road environment ahead of the host vehicle based on the first road environment information acquired by the first road environment information acquisition means and the second road environment information acquired by the second road environment information acquisition means. Environmental recognition means,
A driving control system for an automobile, comprising:   The first road environment information acquisition means is road sign recognition means for recognizing a road sign of a toll gate, and the second road environment information acquisition means is a toll gate recognition means for recognizing a toll gate guide plate or road sign. The vehicle travel control system according to claim 1, wherein:   The road environment recognition means includes a distance estimation means for estimating a distance to a toll gate according to information on a road sign recognized by the road sign recognition means, an estimated distance estimated by the distance estimation means, and the toll gate 3. The vehicle travel control system according to claim 2, comprising distance correction means for detecting a distance to the toll gate according to information on a toll gate guide plate or road marking recognized by the recognition means. .   4. The vehicle travel control system according to claim 1, further comprising: a control unit that controls the host vehicle in accordance with a road environment recognized by the road environment recognition unit. 5. .   The informing device according to any one of claims 1 to 3, further comprising: an informing unit that informs an occupant of a traveling state of the vehicle according to a road environment recognized by the road environment recognizing unit. Automobile running control system.   The said 1st road environment information acquisition means and the said 2nd road environment acquisition means utilize the image imaged by the at least 1 imaging device, The any one of Claim 1 thru | or 3 characterized by the above-mentioned. The vehicle running control system described in 1. Distance prediction means for calculating a predicted distance from the vehicle position to the feature according to the GPS signal and the map database;
When the predicted distance is within a predetermined distance, it recognizes a sign to a feature existing at the destination of the vehicle, and determines the feature from the vehicle position according to information on the sign to the feature and the predicted distance. A sign recognition means for calculating an estimated distance to
Feature recognition means for recognizing the feature when the estimated distance is within a predetermined distance and calculating a correction distance from the vehicle position to the feature according to the feature information and the estimated distance;
When the correction distance is within a predetermined distance, control means for controlling the vehicle as having arrived at the feature;
A vehicle control device comprising:   The vehicle control device according to claim 1, wherein the control unit notifies the occupant of the traveling state of the vehicle according to the recognized sign and the feature.   The vehicle control apparatus according to claim 7 or 8, wherein the sign recognition means recognizes a sign to the feature from an image captured by one or more cameras mounted on the vehicle. .   The vehicle control device according to claim 7 or 8, wherein the feature recognition means recognizes the feature from an image captured by one or more cameras mounted on the vehicle.   11. The sign recognition means recognizes a sign to a toll gate, and the feature recognition means recognizes a toll gate guide board or road marking. The vehicle control device described in 1.   The sign recognition means recognizes a sign to a branch road configured as a rampway, interchange, or parking area on a highway, and the feature recognition means recognizes a guide board on the branch road. The vehicle control device according to any one of claims 7 to 10.   8. The sign recognition means recognizes a sign indicating “forward sharp curve”, and the feature recognition means recognizes a curve curvature and a relaxation curve before the curve from a white line on the road. The vehicle control device according to claim 10.   The sign recognition means recognizes a sign indicating "front crossing", and the feature recognition means recognizes a crossing breaker and a red flashing light. The vehicle control device according to claim 1.   For the blue background guide plate in which the direction and distance indicating the intersection are written in white characters, the sign recognition means recognizes the guide board before the intersection, and the feature recognition means uses the guide board immediately before the intersection. The vehicle control device according to any one of claims 7 to 10, wherein the vehicle control device is recognized.

Images